It’s possible that I shall make an ass of myself. But in that case one can always get out of it with a little dialectic. I have, of course, so worded my proposition as to be right either way (K.Marx, Letter to F.Engels on the Indian Mutiny)
Antimicrobial-resistant Salmonella poses severe challenges to global food safety and public health.
Biofilms formed by Salmonella on food and food-processing equipment are difficult to eliminate with conventional disinfection methods.
Researchers have discovered an alternative method using the bacteriophage W5, which specifically targets Salmonella, paving the way for novel phage-based disinfectants.
Washington, D.C.—Researchers from China have identified a novel bacteriophage that offers a highly promising “green” biocontrol solution against foodborne Salmonella. The study was published in Applied and Environmental Microbiology, a journal of the American Society for Microbiology.
This study was conducted to address the severe challenges posed by antimicrobial-resistant Salmonella to global food safety and public health. Conventional disinfection methods often fail to effectively eliminate the stubborn biofilms formed by Salmonella on food and food-processing equipment surfaces, and the overuse of antibiotics has further accelerated the emergence of drug-resistant strains. There is an urgent need to develop novel, targeted and sustainable alternative antibacterial strategies. Bacteriophages, viruses capable of specifically lysing bacteria, offer a highly promising solution.
In the new study, the researchers isolated bacteriophages that target Salmonella from wastewater and selected the most effective one, phage W5, from multiple candidates. The researchers characterized W5's morphology, stability under various conditions, growth kinetics and genomic sequence to confirm its efficacy and safety. They also evaluated W5's ability to reduce Salmonella and disrupt biofilms on foods (milk, meat, eggs) and food-contact surfaces under realistic storage conditions.
“We discovered a safe and highly effective natural virus (bacteriophage W5) that functions like a precision-guided missile, capable of eliminating harmful Salmonella on various foods and packaging materials, showing great potential as a novel guardian for food safety,” said corresponding study author and professor Huitian Gou from the College of Veterinary Medicine, Gansu Agricultural University in Lanzhou, China. “The research demonstrates that W5 can efficiently lyse planktonic bacteria and eradicate biofilms with high specificity. Genomic analysis further confirms its safety profile, as it lacks virulence and antibiotic resistance genes.”
The researchers say the findings establish a solid foundation for developing novel phage-based disinfectants or preservatives, opening an innovative pathway to combat antibiotic resistance and enhance food safety. As a natural biological entity, phage W5 offers a "green" solution for decontamination, aligning with consumer demand for clean-label products and sustainable production methods. It leaves no harmful chemical residues on food or in the environment.
“We firmly believe that phage W5 holds immense potential for seamless integration across the entire from farm to fork supply chain. It can be incorporated into multiple critical stages—for instance, as a feed additive in livestock farming, a surface disinfectant in meat processing plants, or even a preservative spray for fresh produce at the consumption end,” Gou said. “We eagerly look forward to collaborating with industry partners to translate this effective green solution from the laboratory to the market, working together to safeguard food safety.”
###
The American Society for Microbiology is one of the largest professional societies dedicated to the life sciences and is composed of over 38,000 scientists and health practitioners. ASM's mission is to promote and advance the microbial sciences.
ASM advances the microbial sciences through conferences, publications, certifications, educational opportunities and advocacy efforts. It enhances laboratory capacity around the globe through training and resources. It provides a network for scientists in academia, industry and clinical settings. Additionally, ASM promotes a deeper understanding of the microbial sciences to all audiences.
Journal
Applied and Environmental Microbiology
Sunday, March 08, 2026
UH OH
With Evo 2, AI can model and design the genetic code for all domains of life
The largest foundation model for biology to date is now published in the journal Nature
The DNA foundation model Evo 2, first released in February 2025 as a preprint, is now published in the journal Nature. Trained on the DNA of over 100,000 species across the entire tree of life, Evo 2 can identify patterns in gene sequences across disparate organisms that experimental researchers would need years to uncover. The machine learning model can accurately identify disease-causing mutations in human genes and is capable of designing new genomes that are as long as the genomes of simple bacteria.
Evo 2 was developed by scientists from Arc Institute and NVIDIA, convening collaborators across Stanford University, UC Berkeley, and UC San Francisco. The model's code is publicly accessible from Arc's GitHub, and is also integrated into the NVIDIA BioNeMo framework, as part of a collaboration between Arc Institute and NVIDIA to accelerate scientific research. Arc Institute also worked with AI research lab Goodfire to develop a mechanistic interpretability visualizer that uncovers the key biological features and patterns the model learns to recognize in genomic sequences. The Evo team has shared its training data, training and inference code, and model weights, making it the largest-scale, fully open source AI model to date.
Building on its predecessor Evo 1, which was trained entirely on single-cell genomes, Evo 2 is the largest artificial intelligence model in biology to date, trained on over 9.3 trillion nucleotides—the building blocks that make up DNA or RNA—from over 128,000 whole genomes as well as metagenomic data. In addition to an expanded collection of bacterial, archaeal, and phage genomes, Evo 2 includes information from humans, plants, and other single-celled and multi-cellular species in the eukaryotic domain of life.
"Our development of Evo 1 and Evo 2 represents a key moment in the emerging field of generative biology, as the models have enabled machines to read, write, and think in the language of nucleotides," says Patrick Hsu, Arc Institute Co-Founder, Arc Core Investigator, an Assistant Professor of Bioengineering and Deb Faculty Fellow at University of California, Berkeley, and a co-senior author on the paper. "Evo 2 has a generalist understanding of the tree of life that's useful for a multitude of tasks, from predicting disease-causing mutations to designing potential code for artificial life. We're excited to see what the research community builds on top of these foundation models."
Evolution has encoded biological information in DNA and RNA, creating patterns that Evo 2 can detect and utilize. "Just as the world has left its imprint on the language of the Internet used to train large language models, evolution has left its imprint on biological sequences," says co-senior author Brian Hie, an Assistant Professor of Chemical Engineering at Stanford University, the Dieter Schwarz Foundation Stanford Data Science Faculty Fellow, and Arc Institute Innovation Investigator in Residence. "These patterns, refined over millions of years, contain signals about how molecules work and interact."
Evo 2 was trained for several months on the NVIDIA DGX Cloud AI platform via AWS, utilizing over 2,000 NVIDIA H100 GPUs and bolstered by collaboration with NVIDIA researchers and engineers. The model can process genetic sequences of up to 1 million nucleotides at once, enabling it to understand relationships between distant parts of a genome. Achieving this technical feat required the research team to reimagine how an AI model could quickly ingest and make inferences about this scale of data. The resulting AI architecture, called StripedHyena 2, enabled Evo 2 to be trained with 30 times more data than Evo 1 and reason over 8 times as many nucleotides at a time.
The model already shows enough versatility to identify genetic changes that affect protein function and organism fitness. For example, in tests with variants of the breast cancer-associated gene BRCA1, Evo 2 achieved over 90% accuracy in predicting which mutations are benign versus potentially pathogenic. Insights like this could save countless hours and research dollars needed to run cell or animal experiments, by finding genetic causes of human diseases and accelerating the development of new medicines.
In addition to genetic analysis, Evo 2 could be useful for engineering new biological tools or treatments. "If you have a gene therapy that you want to turn on only in neurons to avoid side effects, or only in liver cells, you could design a genetic element that is only accessible in those specific cells," says co-author and computational biologist Hani Goodarzi, an Arc Core Investigator and an Associate Professor of Biochemistry and Biophysics at the University of California, San Francisco. "This precise control could help develop more targeted treatments with fewer side effects."
The research team envisions that more specific AI models could be built with Evo 2 as a foundation. "In a loose way, you can think of the model almost like an operating system kernel—you can have all of these different applications that are built on top of it," says Arc's Chief Technology Officer Dave Burke, a co-author on the paper. "From predicting how single DNA mutations affect a protein's function to designing genetic elements that behave differently in different cell types, as we continue to refine the model and researchers begin using it in creative ways, we expect to see beneficial uses for Evo 2 we haven't even imagined yet."
In consideration of potential ethics and safety risks, the scientists excluded pathogens that infect humans and other complex organisms from Evo 2's base data set, and ensured that the model would not return productive answers to queries about these pathogens. Co-author Tina Hernandez-Boussard, a Stanford Professor of Medicine, and her lab members assisted the team to implement responsible development and deployment of this technology.
"Evo 2 has fundamentally advanced our understanding of biological systems," says Anthony Costa, director of digital biology at NVIDIA. "By overcoming previous limitations in the scale of biological foundation models with a unique architecture and the largest integrated dataset of its kind, Evo 2 generalizes across more known biology than any other model to date — and by releasing these capabilities broadly, Arc Institute has given scientists around the world a new partner in solving humanity's most pressing health and disease challenges."
Arc Institute is an independent nonprofit research organization based in Palo Alto, California, that aims to accelerate scientific progress and understand the root causes of complex diseases. Arc's investigators are supported by long-term funding and freedom to pursue bold ideas. Its Technology Centers leverage multi-omics, genome engineering, and cellular, mammalian and computational models to advance discoveries at the intersection of biology and artificial intelligence. Founded in 2021, Arc partners with Stanford, UC Berkeley, and UCSF.
A Caltech-led team of biochemists has homed in on an underexplored small transporter called MurJ that is a vital part of the pathway bacteria use to build their chain-mail-like cell wall. An essential component of the cell wall, called peptidoglycan, provides the strength that allows bacteria to resist pressure. Using advanced tools, the scientists have determined the common mechanism used by three different bacteria-killing viruses to block MurJ from doing its job. The findings reveal a novel target for designing new antibiotics.
Here, three distinct phage Sgl proteins lock the flippase MurJ in an outward-facing state, providing a template for antibiotic discovery.
A paper about the new work was published online in the journal Nature on February 25. The lead author of the paper is Yancheng Evelyn Li, a graduate student in the lab of Bil Clemons, the Arthur and Marian Hanisch Memorial Professor of Biochemistry at Caltech, who is the corresponding author.
"Evolution is powerful, and in bacteria, resistance to antibiotics develops quickly. This means that we now deal with bacteria that are resistant to all the medicines that we have," Clemons says. "In the USA alone, tens of thousands of people die every year from antibiotic-resistant bacterial infections, and that number is rising rapidly. We need new antibiotics to combat this."
Scientists have long been interested in the cellular pathway that builds peptidoglycan, aptly known as the peptidoglycan biosynthesis pathway, as an antimicrobial target. "Peptidoglycan is a unique feature of bacteria, and that makes it an attractive antibiotic target," Clemons says.
Many details of the peptidoglycan biosynthesis pathway are known and have been leveraged as targets for antibiotics. The first pharmaceutical, discovered by Alexander Fleming in the middle part of the last century, was the antibiotic penicillin. It and its derivatives, such as amoxicillin, target a late step in this pathway to kill bacteria.
In bacteria, three key proteins—MraY, MurG, and MurJ—facilitate the transfer and transport of peptidoglycan's building blocks from within the cell across the inner membrane barrier. If any of the three proteins fail, peptidoglycan cannot be made, and bacteria die, making them exciting targets for antibiotic discovery. Scientists know a lot about these proteins, but, as noted by Clemons, many basic mechanistic questions remain unanswered.
While the benefits of inhibiting these proteins are clear, there are currently no medicines that target them. However, Clemons says, "We do know that we can find small molecules, either derived from nature or synthesized in chemical libraries, that will inhibit these proteins. Excitingly, recent discoveries have shown that bacteriophages have figured out how to target this pathway."
The survival of viruses that target bacteria, called bacteriophages, or phages, depends on their ability to enter the bacterial cell, make copies of themselves, and then leave to spread as widely as possible. "Getting back out means that they have to get past the peptidoglycan layer. Because it acts like chainmail, the phages get stuck if they can't break through it," Clemons explains.
The Clemons lab has turned some of its focus to single-stranded DNA and RNA phages, tiny phages with small genomes that require simple methods for killing bacteria. In 2023, the lab published a paper in Science about one such phage, φX174, that has a long history at Caltech.
The weapons these small phages use to kill bacteria are protein antibiotics called single-gene lysis proteins, or Sgls (pronounced like “sigils”). Most recently, Li and Clemons have focused on Sgls that target MurJ for antibiotic discovery. MurJ is a flippase, a protein that "flips" peptidoglycan building blocks across the cellular membrane so they can be used to build the peptidoglycan chain. Collaborators had already shown that two Sgls, SglM and SglPP7—which are unrelated and produced by two different phages—both cause bacterial death by inhibiting MurJ.
In the current work, Li used Caltech's Beckman Institute Biological and Cryogenic Transmission Electron Microscopy (Cryo-EM) Resource Center to reveal how these two Sgls inhibit MurJ's flipping activity. Flippases, like MurJ, work by alternating the access of the molecules they transport between the two sides of the membrane without ever making an opening in the membrane. For MurJ, binding of the peptidoglycan precursor within the cell triggers a structural change that effectively moves the molecule outside the cell. Li found that both Sgls bind to a groove in the flippase that prevents the protein from making these structural changes.
"It is clear that both of these Sgls bind to MurJ in an outward-facing conformation, locking it into this position," Li says. That is exciting to researchers because the outward-facing conformation of MurJ is accessible to the surrounding environment. In theory, that makes it easier to target with antibiotics than an internal-facing conformation.
Clemons says the discovery is shocking for another reason. "These peptides, which have no evolutionary links to each other, have both figured out how to target MurJ in a very similar way. These are two examples of convergent evolution, in which different evolutionary paths arrive at the same solution. We were surprised!"
The researchers add that because viruses evolve rapidly, there is likely an endless supply of phages that will all have Sgls. Because phages are easy to find, mining these viral genomes can lead to new biological discoveries and new antibiotic targets. In the Nature paper, the scientists did just that with a new phage. Working with a collaborator, they identified a new Sgl, called SglCJ3 (from a genome sequence that is predicted to be a phage and is called Changjiang3), for cryo-EM analysis. Li resolved the structure of SglCJ3 bound to MurJ and found that it also binds in the same outward-facing conformation of MurJ.
"This is a third genome that evolved a distinct peptide to inhibit the same target in a similar way," Clemons says. "It is the first strong evidence that evolution identifies MurJ as a great target for killing bacteria, which means we should follow evolution's lead and develop therapeutics that target MurJ. This demonstrates the power of basic biology to help us solve problems in medicine. Our path is set on leveraging Sgl discovery, and we hope to continue to be supported to turn these concepts into realities."
The paper is titled "Convergent MurJ flippase inhibition by phage lysis proteins." Along with Clemons and Li, additional authors are Caltech graduate student Grace F. Baron; and Francesca S. Antillon, Karthik Chamakura, and Ry Young of Texas A&M University. The work was supported by the Chan Zuckerberg Initiative, the National Institutes of Health, the G. Harold and Leila Y. Mathers Foundation, and the Center for Phage Technology at Texas A&M, jointly sponsored by Texas A&M AgriLife.
A Caltech-led team of biochemists has homed in on an underexplored small transporter called MurJ that is a vital part of the pathway bacteria use to build their chain-mail-like cell wall. The scientists have determined the common mechanism used by three different bacteria-killing viruses to block MurJ from doing its job.
Here, MurJ from E. coli transitions from an inward to an outward-facing state, where it is locked by a Sgl protein from one of these bacteria-killing viruses.
